The link "vaxgen2.svg" at the bottom of this page is an animated VaxGen sensitivity analysis of the difference in the probability of initiating antiretroviral therapy (placebo minus vaccine) among nonwhites who would have been infected regardless of treatment assignment. This analysis is an extension of Figure 4 in "Sensitivity analyses comparing time-to-event outcomes only existing in a subset selected postrandomization and relaxing monotonicity" by Shepherd BE, Gilbert PB, and Dupont CT (submitted) to multiple time points after HIV-infection. Please refer to that paper for details.

In short, each slide shows a sensitivity analysis of SCE(t), varying phi, beta_0, and beta_1. The time after HIV-infection, t, is given in months and varies from 1 to 24. Contours represent the estimated SCE(t) at a given level of sensitivity parameters. Shaded regions correspond to those sensitivity parameters where the Wald-based 95% confidence intervals for SCE(t) do not contain 0. Estimates in the dark-shaded regions imply that among those who would have been infected regardless of treatment assignment, vaccination lowered the probability of starting antiretrovirals by t months of infection diagnosis (was beneficial), whereas estimates in the light-shaded regions imply that vaccination increased the probability of starting antiretrovirals within t months (was harmful).

Please view using Firefox. The buttons at the bottom of the animation allow the user to start, pause, and fast-forward or rewind the animation to the desired month t.

vaxgen2.svg
Topic revision: r5 - 07 Jun 2010, CharlesDupont
 

This site is powered by FoswikiCopyright © 2013-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback